Lenis and Fortis in IE

[Glen Gordon]

MIGUEL (on the topic of potential Greek borrowings of Anatolian):
>Probably because Anatolian t was aspirated, or sounded aspirated
>to the Greeks.

Alright so Anatolian *t was /t<h>/? I've been thinking about your idea
on fortis and lenis stops existing in IE (which is eerily similar to an
idea I had about Pre-IE which has been evolving for six monthes or more
now). I'm already prone to accept your idea in some form.

About Pre-IE, I was considering a while back that although it seems
unlikely to me that IE had ejectives (since their glottalic quality
seems conveniently to disappear with almost no trace), I considered an
alternative idea - that Pre-IE in fact had lenis and fortis stops that
were the result of a shift from earlier ejectives (This is in connection
with Uralic actually but I will digress).

However, I figure that the _fortis_ stops are what Gam. call the
"glottalic" and would thus correspond to the *d/*g series. The lenis
consonants are the voiced aspirate (*dh) and voiceless plain stops (*t)
of old, all of which I had interpreted as voiceless and I believe I
mentioned this earlier on the list.

   Pre-IndoEtruscan (Stage 1)     *t<h>    *t?     *t
   Pre-IndoEtruscan (Stage 2)     *t<h>    *t:     *t

           IndoEtruscan           *t<h>    *t      *tT
                     Etruscan     t        t       th
                     IE           *t<h>    *t      *tT

                  Traditional     t       d      dh

In this scenario, Hittite medial geminates would correspond to the
aspirate stops in opposition to a merged set corresponding to both the
voiceless affricates (*dh, *gh, *bh) and the voiceless inaspirates (*d,
*g).

Maybe fortis/lenis distinctions in IE proper might have validity but I
suspect a more evolved situation from this distinction (cf *t > IE *tT
(*dh)). Sanskrit voiced aspirates would have evolved from voiceless
affricates as are found in original voiceless form after mobile *s- and
in all environments in Greek. The original voicelessness of the "voiced
inaspirates" can be found in Germanic and in the Latin initial stops for
example.

What I like about this idea is that the "fortis" consonants correspond
exactly to an earlier "ejective" set that thus explains the lack of *b
(which would have been *p? > *p: and then uniquely merging with *p - I
guess lips are too weak to make the distinction :) The glottalic theory
therefore still supplies the best explanation so far of lack of *b but
there is no need for supporting ejectives in actual IE proper.

[ Moderator's comment:
  Modern Estonian has a three-way opposition in the obstruents, lenis ~ fortis
  ~ geminate (fortis) (traditionally "short" ~ "long" ~ "overlong").  According
  to Ilse Lehiste, a native speaker, the word for "Help!" is [ap:i]--and the
  obstruent may be held for some time.
  --rma ]

Optionally, assuming only that you accept Etruscan and IE relationship,
a correspondance can be seen between the two as shown above. Note if
Etruscan <pi> matches IE *bhi then it suggests to me that a loss of *p:
was very early and affects both languages.

Suggestion:
  Pre-IEtr [*t<h>, *t, *tT] >
           [*t<h>/*t,  *tT] >
       Etr [t,         th]

  BUT...

  Pre-IEtr [*p<h>,  -, *pf] >
           [*p<h>, *p,   -] >        <--- !!!
           [*p<h>/*p,  NIL] >
       Etr [p,         NIL]    (I'm aware Etr has <ph> already :)

Ist es cool? I'm sleepy right now so when you respond, be gentle.

--------------------------------------------
Glen Gordon
glengordon01 at hotmail.com

Kisses and Hugs
--------------------------------------------